Zpětné testování modelů pro odhad rizika na bázi Value at Risk

Abstract

The aim of the thesis is Backtesting of Value at Risk based models at a confidence level of 99%. The thesis is divided into five parts, the first of which is the introduction and the last conclusion. The second chapter focused on the methodological part, which described the financial market regulation and supervision, and risk and it‘s categorization and subsequently given stochastic processes. In the third chapter describes the Value at Risk and backtests, which are the Basic frequency test, Basel traffic light and Kupiecuv unconditional test. The fourth chapter is devoted to the application of Value at risk at levels of 0.1%, 1%, 2.5% and 5% and then backtests are calculated on a normal and variance gamma probability distribution. Time horizons are holding a day, 10 days and 20 days.